REDUCING THE EFFECTS OF REVERBERATION IN RADIO BROADCASTING USING AN INTELLIGENT ECHO CANCELLATION TECHNIQUE

ABSTRACTIn this study, an intelligent technique for echo cancellation in radio broadcasting systems was proposed. Oscillations(reverberations) occur in the On-Channel Repeater(OCR) located in the transmission segments of Digital Audio Broadcasting(DAB) terrestrial systems as a result of echoes. This affects the stability of the radio broadcasting system, it increases transmission processing … See more

ABSTRACTIn this study, an intelligent technique for echo cancellation in radio broadcasting systems was proposed. Oscillations(reverberations) occur in the On-Channel Repeater(OCR) located in the transmission segments of Digital Audio Broadcasting(DAB) terrestrial systems as a result of echoes. This affects the stability of the radio broadcasting system, it increases transmission processing delay, weakens the radio broadcast signals, it limits the coverage area of the radio broadcast and increases the difficulty of demodulating the broadcast signal. In this work the Hyper Radial Basis Function (HRBF) neural networks was modeled for the echo cancellation. The HRBF was modeled and trained as a filter for estimating the echo path for effective cancellation. Simulations were used to validate the work. The Radio Nigeria repeater station at Milikin Hill Ngwo Enugu Nigeria was used as a test bed, the radio broadcasting system and the echo cancellation scheme were implemented in MATLAB/SIMULINK simulation environment. Results from evaluation carried out showed that, with respect to the existing Least Mean Square based Finite Impulse Response (FIR) filter echo cancellation technique, the proposed intelligent echo cancellation scheme realized a 28.54% increment in echo cancellation, a 35.5% reduction in radio broadcast transmission processing delay and a 4.06% increment in echo cancellation rate. Furthermore,the transmission processing delay with the proposed scheme exceeded the required delay threshold be 6.47% whereas with the existing LMS based FIR filter scheme, it exceeded the delay threshold by 60.10%. Compared to the LMS based FIR filter echo canceller, the proposed scheme is less likely to affect the demodulation of radio broadcast signal since it is processing delay exceeded the required delay threshold by a very negligible amount.

See less